00:00:00 --- log: started forth/04.07.23 00:25:03 --- quit: tgunr (Read error: 104 (Connection reset by peer)) 00:26:35 --- join: tgunr (~davec@vsat-148-63-4-106.c001.g4.mrt.starband.net) joined #forth 00:36:58 * Tomasu is away: dlrrp 00:37:03 --- nick: Tomasu -> TomasuDlrrp 00:45:26 --- join: arke (arke@melrose-251-251.flexabit.net) joined #forth 00:46:01 HI! 00:46:43 Terve, arke! 00:59:29 privet :) 01:06:12 --- quit: jDoctor (Read error: 60 (Operation timed out)) 01:09:43 --- quit: tgunr (niven.freenode.net irc.freenode.net) 01:09:43 --- quit: tathi (niven.freenode.net irc.freenode.net) 01:09:43 --- quit: kuvos (niven.freenode.net irc.freenode.net) 01:09:43 --- quit: TomasuDlrrp (niven.freenode.net irc.freenode.net) 01:11:30 --- join: tgunr (~davec@vsat-148-63-4-106.c001.g4.mrt.starband.net) joined #forth 01:11:30 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 01:11:30 --- join: kuvos (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 01:11:30 --- join: TomasuDlrrp (~moose@S010600045a4c73cc.ed.shawcable.net) joined #forth 01:18:35 --- quit: arke ("Roses are red, Violets are blue, stop smoking my bong, I want some too!") 02:00:04 --- join: qFox (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 02:24:14 --- join: arke (arke@melrose-251-251.flexabit.net) joined #forth 02:24:20 hi qFox 02:24:22 hi everybody 03:10:08 ... 03:10:09 :( 03:10:16 everybody sleeps? :( 03:18:02 --- quit: arke ("Lost terminal") 03:51:31 So where have you gone? 04:11:29 picking flowers 04:20:53 --- join: crc (crc@0-1pool176-13.nas6.philadelphia1.pa.us.da.qwest.net) joined #forth 04:21:33 tathi: I still haven't managed to get non-line blocking input to work under Linux 04:21:39 It segfaults :-( 04:23:36 Ouch 04:24:38 So I'm uploading beta 5 with the existing KEY implementation 04:25:27 http://www.retroforth.org/dev/beta_releases/beta5/ for those who want to try it 04:25:44 The individual ports are now separated, so the source tree is cleaner :-) 04:25:59 I've tested the Linux and Windows ports, the others should work too 04:26:14 Though the makefiles (new) have not been tested yet 04:26:37 No documentation or sample programs included in these betas; those will be separate as well :-) 04:34:47 Dobryjj den'! 04:34:55 Privet, ASau :) 04:35:16 crc, does that mean I need previous 60KB source tar ball? 04:36:06 Good morning ASau 04:36:22 No 04:36:41 The new source trees are much smaller (~9k including a binary for everything except Windows) 04:36:58 Why Apache says rf-freebsd/linux.tar.gz is 9KB? 04:37:02 The example code + scanty docs in the earlier tarballs aren't really essential 04:37:15 That's the size of the linux source tree, compressed 04:37:25 Wah! 04:37:41 The windows port has a lot of additional crap due to the system interface (macros + equates and the like) 04:37:56 My source tree is pretty compact ;-) 04:38:27 About 25k decompressed 04:38:54 For linux/freebsd at least 04:39:26 168k for Windows 04:40:10 About 100k for BeOS, including the binary size (binary is not included for BeOS) 04:40:57 I hope it's not too big still... 04:46:29 I think there's a bug in < 04:50:16 Is < written in Forth? 04:50:29 1 2 < returns FALSE, should it return TRUE? 04:50:46 Inline assembly in source/blocks/000 04:51:22 If it's wrong, I'll just rename it '>' :-) 04:51:47 My only conditionals are '=' and '<' 04:52:21 Because it's easy enough to define <> as ': <> = not' and '>' as ': > swap < ;' :-) 04:52:59 I have a version based on the only 0< . 04:53:09 0< suffice. 04:53:22 I don't have 0< 04:53:47 I decided that it's better to have only 0< 04:53:51 Why? 04:54:05 It's rol; and ax,1; push 04:54:19 The most simple primitive. 04:54:28 Good reason :-) 04:55:24 I still like < though 04:55:36 How do you do '1 2 <' with 0< ? 04:55:39 I've got problem with fasm here. 04:55:55 What's the problem (and what OS are you using)? 04:56:27 BSD 5.2.1, "ELF binary type 0 is not known." 04:56:34 What should that mean? 04:57:22 Did you use 'brandelf' ? 04:57:52 hi all 04:57:53 What's it? 04:58:04 I'm new to BSD. 04:58:14 You want to run the Linux binary for FASM right? 04:58:18 Right. 04:58:26 Ah. 04:58:48 You mean I need install separate support for this. 04:59:33 The Linux support files (I haven't recieved my copy of freebsd yet, so I can't tell you what it's called) 04:59:43 But if you have Linux binary support, do: 04:59:46 brandelf -t Linux fasm 04:59:48 And it should work 04:59:59 Or you can use the LIBC backed version :-) 05:01:30 http://board.flatassembler.net/download.php?id=852 05:02:24 Decompress it, enter source/libc and do: 05:02:32 'gcc fasm.o -o fasm -s' 05:02:36 To run that version 05:02:59 * crc runs the libc version 05:03:09 * crc also maintains the libc version :-) 05:09:19 BTW, it looks yesterday they've released FASM 1.53. 05:09:38 You may wish to update. 05:10:29 I will be updating the libc port over the weekend 05:11:45 What's error 146? 05:11:46 1.53 has a few bug fixes with sections and the like, but nothing that directly affects RetroForth 05:11:53 In FASM? 05:12:15 When you try to compiler or run it? 05:12:42 s/compiler/compile/ 05:15:53 Well, I've managed it. 05:16:07 good 05:16:16 No major problems I hope? 05:16:28 You need insert "brandelf -t FreeBSD rf" into Makefile. 05:16:34 Ok 05:16:46 I'll do that for the final release 05:17:11 * crc makes a note of that in the Forums as well\ 05:17:19 And there's another problem in Makefile I've not realized. 05:17:32 What's that? 05:18:22 That "Error code 146" above. 05:19:08 Is that an error when building RetroForth? 05:19:32 When I "make". 05:20:12 * crc is searching for references to 'error 146' under BSD 05:21:26 It's probably due to my use of GNU Make instead of BSD Make 05:21:45 * crc will fix the BSD makefile as soon as he gets his copy of FreeBSD 05:23:09 No. 05:23:25 That's after FASM: source file not found. 05:23:31 ??? 05:23:39 GNU make does the same: 146. 05:23:45 * crc looks at the FreeBSD source tree 05:24:26 It *looks* ok to me 05:24:36 Can you build it manually: 05:24:37 cd source 05:24:44 cat blocks/* >retroforth.f 05:24:52 fasm freebsd.asm ../bin/rf 05:24:56 cd .. 05:24:57 cd bin 05:25:03 I've done it manually. 05:25:05 chmod +x rf 05:25:08 And it works? 05:25:13 It works. 05:25:16 * crc is confused now 05:25:33 I'm meditating on your Makefile. 05:25:39 The Makefile is nearly identical to the Linux one: the only difference is the name of the file being assembled! 05:26:01 And it worked under Linux until I deleted 'make' 05:26:20 you deleted make? 05:26:27 Yes 05:26:41 I'm preparing to repartition + reinstall Linux and BeOS 05:26:47 I need to make space for FreeBSD :-) 05:27:28 oh, ok. 05:28:04 I've also been experimenting with minimizing the toolchain I use under Linux 05:28:55 minimizing meaning...? 05:29:09 not depending on the most recent version of tools? 05:29:13 using fewer tools? 05:29:32 Fewer tools 05:29:32 Latter, obvioursly. 05:29:46 I've been paring it down for over two years 05:30:15 I try to use the latest versions of the tools, unless they become overly bloated 05:30:36 What things have you stopped depending on? 05:30:49 GCC, X11, and so on 05:31:22 oh. 05:31:38 I mainly use a simple shell (called KISS), I compile C with TinyCC, I use FASM, Nano (sten/te aren't done yet...) 05:31:41 I'm so used to only depending on make and gas (and a few shell tools, usually) 05:32:02 My Linux distro is *just* what I use on a daily basis 05:32:16 Maybe 30-40 tools total 05:32:25 that's pretty cool. 05:33:33 I think so 05:33:46 It'll all fit on a floppy soon 05:33:59 crc, I've rebuild your LIBC patch with newer FASM. 05:34:31 But I've not checked what files have changed since FASM 1.52. 05:35:05 The files that changed are only the ones in source/ 05:35:16 The source/libc/ files are static 05:36:42 Ah, no, that doesn't go. It's erroneous. 05:38:03 As I said, I'll update it on Saturday (I don't have time to debug it right now) 05:38:13 It shouldn't take too long though 05:38:40 I understand. 05:40:11 Hopefully I'll have my copy of FreeBSD by then as well 05:45:28 --- join: lalalim_ (~lalalim@p508AA3AA.dip.t-dialin.net) joined #forth 06:02:16 --- quit: lalalim (Read error: 110 (Connection timed out)) 06:05:52 --- join: Topaz (~top@spc1-horn1-6-0-cust217.cosh.broadband.ntl.com) joined #forth 06:10:22 --- quit: tathi ("leaving") 06:25:21 Dobryjj vecher, Topaz! 06:26:19 er, hi :P 06:28:14 I don't remember, what are you writing in Forth? 06:29:09 i was writing a AVR FORTH 06:29:17 which kinda works, although i haven't yet thought of anything to do with it 06:29:20 (though I may be about to) 06:30:47 "Write your own Forth" community. 06:31:45 How many of us have written our own Forths? 06:31:56 --- join: FlamingRain (~Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth 06:32:00 Hello FlamingRain 06:43:08 Time for bed... 06:43:21 Goodnight all! 06:43:24 Night crc 06:43:30 And hi, FlamingRain :) 06:43:31 --- quit: crc (Client Quit) 06:43:39 heh 07:07:24 --- quit: Robert ("Shutting down the computer to avoid HAXXORZ.") 07:10:59 --- join: snowrichard (richard@adsl-068-209-159-248.sip.shv.bellsouth.net) joined #forth 07:13:59 God kvaell, snowrichard! 07:14:16 good morning to you. 07:14:53 --- quit: FlamingRain ("Leaving") 07:15:29 I have been trying to read intel's Pentium 4 manuals. 07:16:17 And what's new there? 07:16:30 Anything interesting? 07:32:49 --- join: nleavrh (~julian@205.241.56.30) joined #forth 07:42:38 --- join: Robert (~snofs@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 07:51:29 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 08:02:28 --- join: wossname (wossname@HSE-MTL-ppp61452.qc.sympatico.ca) joined #forth 08:15:12 --- quit: qFox ("meh ircii's netsplit detection appearantly sux ^^") 08:21:07 --- quit: wossname (":dz") 08:21:48 --- quit: tathi ("leaving") 08:29:36 --- quit: kuvos (Read error: 54 (Connection reset by peer)) 08:30:44 --- join: kuvos (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 08:32:13 --- quit: snowrichard ("ChatZilla 0.9.52B [Mozilla rv:1.6/1]") 08:33:46 [17:36:56] Since 03/16/2004: 08:33:47 [17:36:56] System Availability: 99.9922% 08:34:00 ^^ 08:35:43 --- join: qFox (~C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 08:37:01 --- join: kc5tja (~kc5tja@66-74-218-202.san.rr.com) joined #forth 08:39:37 Dobryjj vecher, kc5tja! 08:40:33 re 08:40:58 Need to head to Oceanside (about 50km north) soon. 08:41:37 Have you any experience in C metaprogramming? 08:42:02 You mean the C pre-processor? 08:42:23 I don't know another "standard" way. 08:42:42 There is no other standard way except the C pre-processor, and it is woefully limited in what it can do. 08:42:58 I generally tend to avoid it like the plague myself. 08:43:14 I understand. 08:43:41 I doubt GNU folk accept Forth code ever. 08:44:34 As a scripting language at best, but even then, they'll push Guile (their Scheme implementation). 08:44:39 if they would, they probably would've done so by now, dont ya think 08:44:40 I've found terrible code, that's equivalent to: >r um* r> um/mod drop 08:45:24 That's a scaling operation, yes? 08:45:53 It's multiplication within finite ring. 08:46:06 x -> (x*a) mod m 08:46:17 OK, I got the order of results for /MOD backwards. 08:47:07 Actually, I've got code of this kind: 08:47:31 x -> (a (x mod q) - r (x div q)) mod m 08:47:57 a, r, q, m are hard-coded constants. 08:48:17 While actually, q = m div a, r = m mod a. 08:48:55 I don't know what all that is supposed to do. 08:49:22 It's widely used in random number generators. 08:50:00 I'm hacking one probabilistic algorythm. 08:50:06 * kc5tja is finally narrowing in on a TTA processor register set for the Kestrel. 08:51:19 I think, how far can I dig with cpp, so that GNU accepted patches. 08:51:39 ...accept... 08:52:12 They've managed to release lots of unreadable code. 08:52:53 Precisely why I advocate a language like Oberon -- languages should enforce readability, because it's patently obvious that humans won't. 08:53:45 I'd liked if this library was written in ML or LISP. 08:53:59 Forth is kind of weird in a sense -- it is very readable provided you are aware of context. Fortunately, most local definitions are used close together, and therefore provides sufficient context to be readable. 08:54:13 They could avoid this mathematical trick. 08:54:51 I can't even follow the trick. 08:55:18 Well, assume we have 32 bits wide cell. 08:55:32 Double cell is 64 bits wide. 08:55:46 x -> (x*a) mod m is 08:56:06 >r um* r> um/mod drop ( x a m -- x' ) 08:56:24 You see you have double cell in the middle. 08:56:35 sure 08:56:45 It looks, C can't boast with double cell. 08:56:52 So they do this: 08:57:11 x = (x mod q) + (x div q) 08:57:32 ax = a (x mod q) + a (x div q) 08:58:26 Wait a minute. Shouldn't x = q((x div q) + (x mod q)) ? 08:58:37 Yes. 08:58:43 ax = a (x mod q) + aq (x div q) 08:59:00 While x = (x mod q) + q (x div q) 08:59:07 Thanks. 08:59:32 n/p 08:59:44 ax = a (x mod q) + aq (x div q) - m (x div q), *in sense of finite ring* 08:59:45 Well, I need to get cruising. 09:00:07 (ax) mod m = a (x mod q) + (aq - m) (x div q) 09:00:45 r = m - aq is taken to be m mod a. 09:01:02 Just to have the smallest possible numbers. 09:01:52 You may have to add m to correct result, but that has less matter. 09:02:30 So you've done all arithmetic operations in one cell. 09:05:11 The problem is you've done one more multiplication and you have to correct result conditionally. 09:08:54 --- nick: TomasuDlrrp -> Tomasu 09:08:56 * Tomasu is back (gone 08:31:59) 09:25:07 --- quit: kc5tja ("leaving") 09:31:07 --- nick: Tomasu -> Vadar 09:31:25 --- nick: Vadar -> Tomasu 09:39:22 --- quit: Robert ("brb, new kernel") 10:00:07 --- quit: qFox (Read error: 104 (Connection reset by peer)) 10:00:27 --- quit: kuvos (Read error: 104 (Connection reset by peer)) 10:01:44 --- join: qFox (~C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 10:36:45 --- quit: SDO (Read error: 104 (Connection reset by peer)) 10:37:52 --- join: SDO (~SDO@67-23-111-213.clspco.adelphia.net) joined #forth 10:47:37 --- join: Robert (~snofs@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 10:51:13 Happy kernel, Robert! 10:51:28 Thanks a lot... 10:51:38 2.6.?? 10:51:49 Or 5.3.?? 10:52:47 2.4.26 10:53:33 Hm. 10:53:35 Why? 10:54:48 --- quit: Robert ("OK - fuck grsec - I'm back to good old 2.4.26. Feel free to HAXX me.") 10:55:33 hmm 10:55:49 notify does not work with *@foo.bar right? 10:57:28 Sorry, I don't know. 11:01:50 --- join: Robert (~snofs@c-bf5a71d5.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth 11:02:33 Happy kernel once again. 11:02:38 Why not 2.6? 11:03:59 I tried it once on this computer, but it had some problems. 11:04:11 But I run 2.6.7 on the other computer. 11:08:59 Maybe, you've selected wrong drivers? 11:14:31 I don't think so. But the oddest thing of all - the computer made strange noises. 11:53:20 --- join: wossname (wossname@HSE-MTL-ppp61452.qc.sympatico.ca) joined #forth 12:04:37 --- quit: ASau ("leaving") 12:31:14 --- join: Sonarman (~matt@adsl-64-160-166-84.dsl.snfc21.pacbell.net) joined #forth 12:34:35 --- quit: I440r (Excess Flood) 12:34:49 --- join: I440r (~mark4@216-110-82-1.gen.twtelecom.net) joined #forth 12:39:06 --- quit: slava (niven.freenode.net irc.freenode.net) 12:50:31 --- join: randolm (wossname@HSE-Hamilton-ppp190651.sympatico.ca) joined #forth 13:03:56 --- quit: wossname (Read error: 110 (Connection timed out)) 13:25:49 --- join: kc5tja (~kc5tja@66-74-218-202.san.rr.com) joined #forth 13:25:54 --- mode: ChanServ set +o kc5tja 13:26:48 hi 13:26:51 re 13:28:34 --- join: slava (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth 13:28:41 re slava 13:52:27 whoo, taking a trip down memmory lane 13:52:52 kinda freaky. cleaning up boxes in which i keep things from my past. letters cards schoolstuff all kind of things 13:53:27 i've found letters from ppl i couldnt even remember, pictures of ppl i knew very well but had to actually think hard for to remember 13:54:00 and god, my mom must've sent me a thousand postcards for the time i didnt live at home. 14:04:58 Heh. 14:06:22 I have very little physical correspondence, and a large deal of it (weight-wise) is now scattered in thw woods. I wrote very few letters between the age of 4 and 17. 14:06:56 First I thought writing itself was new and exciting, then I found that I could put some contents in the letters. 14:13:22 well i obviously dont have the letters i wrote :p 14:13:52 i do believe i lost a few correspondents (aka.. penpals and old friends after a move) due to lack of writing a return letter (eventually), on my side 14:14:08 --- join: T0paz (~top@spc1-horn1-6-0-cust117.cosh.broadband.ntl.com) joined #forth 14:14:33 not much of a writer, penwise 14:14:39 I tend to be very keen on writing replies, if I have something to say (which I do, unless I find that I don't at all get along with the other person). 14:15:51 Personally, I enjoy writing. But somewhere in the process from thought to writing, something goes wrong. I am not a bad writer, and I do come up with lots of things to express, but it often stops there. 14:15:57 well obviously. i tend to POSTPONE it and eventualy forget about it 14:16:00 --- quit: Topaz (Read error: 110 (Connection timed out)) 14:16:09 which is a bad habit :\ 14:16:39 See, ANS is evil. 14:16:43 :p 14:16:55 well i'm just not much of a handwritten writer. 14:16:57 See, THAT was a major flaw in the process from thinking to writing! 14:17:13 i learned to write bad, the pose of my hands and the holding of the pen is wrong :p 14:17:24 on a computer, its virtually no problem 14:17:28 on and on and on and on 14:17:29 :D 14:17:33 I'm quite comfortable with both, even if I (physically) can't write either on a computer keyboard or by hand. 14:17:45 oh i'll write if i have to 14:17:57 i've written 4 .... eh, what are they called in english 14:18:00 notebooks 14:18:00 Same here, but I'm not one of those fast writers. 14:18:03 books 14:18:09 With? 14:18:16 ehr, the papers where you write in your homework 14:18:33 I filled one of those with morse code conversations I had in a couple of months. :) 14:18:53 whatever it is, in the process of learning for my math tests i've written quite a few papers full 14:19:07 but i prefer my keyboard 14:19:17 i think i can actually even type faster then write 14:19:37 And my electronic correspondance would probably fill a book if printed (and IRC is OBVIOUSLY not counted - I once calculated my collected 'writings' on IRC to be rougly equivalent to 1000 pages of text). 14:19:45 Yes, so do I. 14:19:55 although i dont have THAT bad of a handwriting. at least thats what i've been told, from my point of view, it sux, especially after a while 14:19:57 But hand writing is... less disturbing to the thought process. 14:20:13 oh no, i'll top that way 14:20:27 less disturbing? 14:20:31 irc and any other e-writing. whee 14:21:36 randolm: HI FORTHER. As in... I feel like the thoughts can flow more freely when using pen and paper. Computers can be good for printing a letter you have in your head, but just for ideas, I prefer pen and paper. 14:21:56 ah 14:22:12 hm, i have no problem with that 14:22:26 paper is good for code design 14:22:31 what WAS funny though, the smell of certain papers from one of my schools was very familiar 14:22:32 That too. 14:22:34 because that involves broken diagrams 14:22:35 to that specific school 14:22:38 it was so strange 14:22:49 but computers are better for letting go of ideas in writing.. 14:23:18 randolm: No, you just need a lot of paper. :) That's something I like about those, you don't need any organization. 14:23:47 computers allow you to actually read what you wrote, no matter what ;) 14:23:51 It's easy to just draw lines everywhere, a smiling face, and some comments... I know you CAN do it on computers (using FORTH), but it's not the same thing. 14:24:12 That's another matter, I don't always want to read what I write. 14:24:24 haha 14:24:33 computers are good for that 14:24:44 But I can't always use computers, either. 14:24:49 yes you can >:( 14:24:59 No. 14:25:03 well if you dont want to read it 14:25:07 just type non-bland 14:25:08 blind 14:25:45 (look at your keyboard while typing ;) 14:25:57 * Robert returns to his movie, goodbye, letter-loving Forthers. 14:26:07 haha 14:26:12 i hope you bought that movie ;) 14:26:19 Yes...of course I did. 14:26:31 Even though it's (c) 196..something. 14:26:41 ouch 14:26:46 Not really. 14:27:01 Just like with computers, things don't always get better. ;) 14:27:07 But you know that 14:32:18 also i found 2 (complete) decks of nude playing cards from appearantly a time where these were a new fashion ("plastic coated" playing cards), the year being 1977 14:32:24 now how the hell did those end up in those boxes.. 14:45:30 Heh. 15:03:24 hi 15:08:40 ack, sri. I fell asleep 15:08:46 I'ts really, really hot right now. :/ 15:18:51 --- quit: randolm ("^#") 15:20:39 Heh cool... 15:20:46 I have Forthy embedded into an MFC app. ;) 15:20:47 --- nick: T0paz -> Topaz 15:21:24 Heheh 15:21:26 Sweet. 15:21:57 Going to use some forthy script as a test protocol. 15:22:10 Client/server, each talking to each other in Forth. :) 15:22:23 Nothing wrong with that. 15:22:26 --- nick: madwork_ -> madwork 15:22:29 Damn skippy! 15:22:33 Better than XML! 15:22:46 Yep. 15:22:56 huh? 15:23:00 XML sucks ass for use as a protocol. 15:23:06 Archival format, yes. Protocol, no. 15:23:06 oh nvm 15:23:10 It's just too bulky. 15:23:12 Sonarman: :) 15:23:19 :) 15:23:32 EA-IFF-85a, however, WOULD make a nice protocol medium, I think. Especially since it's all binary. 15:23:39 They're using it here for a protocol of sorts, to represent print jobs. Blargh! 15:24:01 1MB XML file to describe 300 print jobs is NOT cool. 15:24:08 what's with using XML for SVG? 15:24:09 Good grief! 15:24:09 Or was that 2MB... 15:25:47 Anyway, it was funny. There was a problem with the XML code that was crashing the printer. Just a little memory leak... it was only about, oh, most of the XML document. Refresh the job list a few times and printer goes bye-bye. 15:25:57 http://www.burched.biz/sydneyx1.html -- GAHH! I have competition! 15:26:01 How can this BEEE?!?!? :D 15:26:26 Whoa... he has PACMAN! :D 15:27:17 HAH! Kestrel is still better than his! BAWAAHAHAH! 15:27:29 Not only am I cheaper, I have 4x the amount of colors on the screen! And more RAM. >:) 15:27:34 Hehe. 15:27:43 And, of course, Forth. 15:27:47 Does his product run any OS? 15:29:14 I don't think so. I think it's just a board and some means of downloading code to it. 15:29:22 It doesn't even have a specific CPU for it. 15:29:25 You download that too. 15:29:26 :) 15:31:12 Hehe. 15:34:14 So, I will be finally firming up on a microprocessor design later this evening. 15:34:44 I have to work from 6PM to 10PM tonight (overtime -- yay!), but once I get home from that, I'll start writing the preliminary specifications for the microprocessor. 15:35:26 Did I miss something? i thought you already had some kind of spec. 15:35:40 I am, unfortunately, going to do away with the hard requirement for 1-cycle instructions. True, all instructions DO execute (firmly) in 1 cycle. But the instruction set is so primitive that things like conditional branches require 2 or 3 instructions to implement. 15:35:52 Hence, some things are going to be multi-cycle. 15:36:14 madwork: You didn't read the Forth logs, didn't you? I have a new CPU architecture in development, a transport-triggered architecture CPU. 15:36:34 The reason for the change is simple: the MISC impinges on Patriot Scientific's and Chuck Moore's patents. 15:36:52 Yes, I've been skimping on my #forth log reading... not enough time. 15:37:01 I did catch something about the patent collision though. 15:37:03 ZISC! 15:37:38 The CPU I'm building is a dual internal bus TTA, and so should be able to achieve 1-cycle Forth primitives for the overwhelming majority of operations. 15:37:50 Memory fetches and more complex stack primitives (e.g., ROT) take two cycles. 15:37:57 Memory stores also take two cycles. 15:39:11 I'm also strongly considering starting off the CPU directly as an asynchronous CPU, and not synchronous. 15:39:24 TTAs appear to be MUCH easier to implement as asynchronous machines, interestingly enough. 15:39:49 --- quit: aum (Connection reset by peer) 15:41:05 Well, that sounds like the perfect plan for you then. 15:41:18 Yeah, but it places a lot of burden on the memory bus. :( 15:41:22 Because I don't have a cache. 15:41:40 But that's OK -- just makes for opportunities to grow the architecture over time. 15:42:15 --- quit: tgunr (Excess Flood) 15:44:43 Hmm... pushing the result of an MFC YESNO MessageBox onto the stack for evaluation just seems so wrong... yet so right. Hahe! 15:44:45 --- join: tgunr (~davec@vsat-148-63-4-106.c001.g4.mrt.starband.net) joined #forth 15:45:05 Visual Forthy 15:46:00 That's about the jist of it. 15:46:09 Same with my game engine's forthy. 15:47:28 I especially like my mauve-while-compiling output. It's a rainbow of Forthy goodness! 15:48:00 my unit tests print code expressions that are being tested, and they are printed in ANSI syntax colored goodness 15:48:17 --- join: aum (~aum@port-204-54-210.fastadsl.net.nz) joined #forth 15:48:58 Colours and Forth... it's like jellybeans and children! 15:50:57 Heheh 15:51:25 * kc5tja wonders whether I should provide compliments to JZ and JP registers (jump if zero and jump if positive respectively) 15:52:02 Can it hurt? 16:08:43 Wow. I have 31 registers defined. That's a lot less than I expected. 16:08:52 (I didn't put in the compliments until I know I need them) 16:19:17 I think my CPU will be the first FPGA CPU to have twin program counters, even though it's not truely multithreaded. 16:36:34 kc5tja, is the new cpu design more complex than the old one? 16:36:47 About the same. 16:36:53 It's architecturally much different though. 16:49:39 --- join: Research (~Research@12-222-128-22.client.insightBB.com) joined #forth 16:53:54 --- quit: SDO ("Vision[0.9.6-0203]: i've been blurred!") 17:06:06 x = (x mod q) + q (x div q) 17:06:23 isn't that the same as x = x? am i missing something? 17:13:22 Curious: when programming in assembly language, do people prefer the "src dst" or "dst src" argument ordering? 17:13:35 Sonarman: That's Asau's point. 17:14:17 He's showing the mathematical derivation for performing a modular multiplication with a "64-bit intermediate result" without resorting 64-bit integers (since C doesn't grok the concept of intermediate results) 17:18:31 i prefer dst src, only because i'm used to it 17:19:25 its whatever you're used to. src dst for me, but you get used to whatever it'll be fast so i dont think its a problem either way 17:20:47 * kc5tja will use src dst layout because it involves less swaps. :D 17:22:18 Well, I have to get going to work. I'll be back in about four and a half hours or so. 17:22:25 --- quit: kc5tja ("THX QSO ES 73 DE KC5TJA/6 CL ES QRT AR SK") 17:34:36 --- quit: Sonarman (Read error: 110 (Connection timed out)) 17:39:49 --- join: kuvos (C00K13S@cp12172-a.roose1.nb.home.nl) joined #forth 17:43:20 --- join: blockhead (default@dialin-633-tnt.nyc.bestweb.net) joined #forth 17:46:11 --- join: Sonarman (~matt@adsl-64-160-165-51.dsl.snfc21.pacbell.net) joined #forth 17:49:59 --- quit: Topaz (Remote closed the connection) 18:09:29 --- nick: blockhead -> bh 18:09:37 --- part: bh left #forth 18:20:44 --- quit: tgunr (Excess Flood) 18:23:12 --- quit: qFox ("meh ircii's netsplit detection appearantly sux ^^") 18:45:48 --- join: tgunr (~davec@vsat-148-63-4-106.c001.g4.mrt.starband.net) joined #forth 18:49:27 --- quit: slava ("Leaving") 18:50:51 --- quit: aum () 19:04:19 --- quit: tgunr (Excess Flood) 19:08:08 --- join: tgunr (~davec@vsat-148-63-4-106.c001.g4.mrt.starband.net) joined #forth 19:11:02 --- join: tucknip (pickroll@dialup-4.153.53.54.Dial1.Atlanta1.Level3.net) joined #forth 19:13:11 --- quit: tgunr (Excess Flood) 19:15:03 --- join: tgunr (~davec@vsat-148-63-4-106.c001.g4.mrt.starband.net) joined #forth 19:23:19 * Tomasu is away: store 19:24:41